Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27308
    Prev
    Next
    INDEX
    Explanations

    references to identification or identity concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_mlp_out
    Hook Layer
    0
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    148
    +0.13
    0.7%
    253
    +0.12
    0.7%
    108
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    7
    +0.13
    0.02
    395
    +0.12
    0.02
    475
    +0.12
    0.02
    Negative Logits
     neither
    -1.65
    .");
    -1.59
    .")
    -1.54
    Ĺ
    -1.53
    .").
    -1.53
     thee
    -1.52
    ");
    -1.52
    Ĥ
    -1.52
    "))
    -1.51
     both
    -1.49
    POSITIVE LOGITS
    iary
    2.29
    nier
    1.69
     era
    1.66
    ifiers
    1.63
    face
    1.63
    ulator
    1.54
    rams
    1.51
    fony
    1.49
    rays
    1.49
    ités
    1.48
    Activations Density 0.016%

    No Known Activations